Re: Optimize date query for large child tables: GiST or GIN? - Mailing list pgsql-performance

From Tom Lane
Subject Re: Optimize date query for large child tables: GiST or GIN?
Date
Msg-id 3387.1274363810@sss.pgh.pa.us
Whole thread Raw
In response to Re: Optimize date query for large child tables: GiST or GIN?  (Matthew Wakeling <matthew@flymine.org>)
Responses Re: Optimize date query for large child tables: GiST or GIN?
List pgsql-performance
Matthew Wakeling <matthew@flymine.org> writes:
> On Wed, 19 May 2010, David Jarvis wrote:
>> extract( YEAR FROM m.taken ) BETWEEN 1900 AND 2009 AND

> That portion of the WHERE clause cannot use an index on m.taken. Postgres
> does not look inside functions (like extract) to see if something
> indexable is present. To get an index to work, you could create an index
> on (extract(YEAR FROM m.taken)).

What you really need to do is not do date arithmetic using text-string
operations.  The planner has no intelligence about that whatsoever.
Convert the operations to something natural using real date or timestamp
types, and then look at what indexes you need.

            regards, tom lane

pgsql-performance by date:

Previous
From: Matthew Wakeling
Date:
Subject: Re: Optimize date query for large child tables: GiST or GIN?
Next
From: Tom Lane
Date:
Subject: Re: merge join killing performance